evaluation of maswrc sample collection, sample analysis, and … · 2020. 3. 3. · evaluation of...
TRANSCRIPT
Evaluation of MASWRC Sample Collection, Sample
Analysis, and Data Analysis
Dr. Allen P. Davis and Ryan Janoch
University of Maryland – College Park
December 22, 2008
Prepared For:
BaySaver Technologies, Inc.
i
EXECUTIVE SUMMARY
In 1972, the federal Clean Water Act (CWA) was enacted and as a result the National
Pollution Discharge Elimination System (NPDES) was created in 1990 to regulate stormwater
discharges. Phase I of the NPDES program covered medium to large municipal stormwater
sewer systems (MS4) in municipalities with greater than 100,000 residents, industrial activities,
and construction activities that disturbed greater than 5 acres. In 2003, small MS4 systems and
construction activities on 1 to 5 acres were added as regulated activities under Phase II of the
NPDES program. Phase II has opened up a market for stormwater treatment devices, as
municipalities and businesses strive to meet NPDES regulations. BaySaver Technologies, Inc.
(BaySaver) of Mount Airy, MD is a stormwater treatment device manufacturer and vendor
whose products are used by site owners to meet the regulatory requirements under the NPDES.
This project was conducted in collaboration with BaySaver as part of the collection and
submission of field testing data of their stormwater treatment devices to New Jersey Corporation
for Advanced Technology (NJCAT). The Mid-Atlantic Stormwater Research Center
(MASWRC), also located in Mount Airy, MD is conducting the field testing of the
BaySeparator™ and BayFilter™ as a treatment train at Richard Montgomery High School in
Rockville, MD (site). The field data being collected will be used for approval of the
BaySeparator™ and BayFilter™, as stormwater treatment devices, under the Tier II protocol
developed by the Technology Acceptance and Reciprocity Partnership (TARP). The University
of Maryland (UMD) is serving as an independent third party, auditing laboratory analysis and
sampling methods, and evaluating the accuracy of the data reporting. In this project, UMD is
evaluating the sampling, testing, and data being reported on three water quality constituents,
suspended sediment, total phosphorus, and turbidity. UMD involvement began in June 2008 and
continued for approximately 6 months, encompassing 17 qualified storm events.
The concentration removal efficiencies of the BaySeparator™ and BayFilter™, as well as
the two devices functioning in series as a treatment train (system), were calculated from water
quality samples collected at the site. Overall the system demonstrated 92% suspended sediment
concentration (SSC), 67% total phosphorus (TP), and 68% turbidity removal efficiency. The
BaySeparator™ had 47% SSC, 27% TP, and 11% turbidity removal efficiency. The BayFilter™
had 86% SSC, 58% TP, and 64% turbidity removal efficiency. Progress in meeting TARP
standards in data collection, analysis, and reporting is documented. Based on this evaluation,
MASWRC monitoring and data analysis methods and techniques appear to be satisfactory.
ii
EXECUTIVE SUMMARY........................................................................................... I
LIST OF ACRO,YMS A,D TERMS ..................................................................... IV
1.0 I,TRODUCTIO,................................................................................................1
2.0 GOALS..................................................................................................................1
3.0 BACKGROU,D...................................................................................................2 3.1 FEDERAL REGULATORY HISTORY ...................................................................2
2.2 LOCAL REGULATORY BACKGROUND ..............................................................2
2.3 STORMWATER TREATMENT .............................................................................4
3.0 THE TREATME,T SYSTEM............................................................................6 3.1 THE SITE ........................................................................................................6
3.2 THE SYSTEM ..................................................................................................7
3.3 SITE CHARACTERIZATION...............................................................................9
4.0 BAYSAVER STORMWATER TREATME,T DEVICES..............................10 4.1 THE BAYSEPARATOR™ ................................................................................10
4.2 THE BAYFILTER™........................................................................................13
5.0 SAMPLI,G ........................................................................................................16 5.1 TARP TIER II PROTOCOL REQUIREMENTS....................................................16
5.2 QAPP...........................................................................................................16
5.3 QUALIFIED EVENTS ......................................................................................17
5.4 EXPERIMENTAL METHODS............................................................................17
5.4.1 Field Sampling ................................................................................17
5.4.2 UMD Sample Collection.................................................................20
5.4.3 Sampling Protocol ...........................................................................20
5.4.4 Analytical Methods .........................................................................21
6.0 RESULTS............................................................................................................25 6.1 PRECIPITATION .............................................................................................25
6.2 RUNOFF/VOLUME BALANCE ........................................................................26
6.3 FIELD FLOW TEST ........................................................................................28
6.3 MONITORING RESULTS .................................................................................31
6.3.1 Constituent Concentrations (Event Mean Concentrations).............32
6.3.2 Removal Efficiency.........................................................................35
6.3.3 Split Samples...................................................................................38
7.0 DISCUSSIO, .....................................................................................................43 7.1 PRECIPITATION .............................................................................................43
7.2 FLOW BALANCE ...........................................................................................43
7.3 MONITORING RESULTS .................................................................................44
7.4 CUMULATIVE MASS......................................................................................44
7.5 NON-QUALIFYING EVENTS ..........................................................................45
7.6 CONCERNS ...................................................................................................45
iii
7.6.1 Concerns Addressed ........................................................................46
7.6.2 Concerns Remaining .......................................................................47
8.0 SUMMARY/CO,CLUSIO, ............................................................................52
REFERE,CES............................................................................................................53
APPE,DIX A – SCOPE OF WORK
APPE,DIX B – TIER II PROTOCOL
APPE,DIX C – QAPP
APPE,DIX D – CHAI, OF CUSTODY
APPE,DIX E – FIELD LOG
iv
LIST OF ACRO,YMS A,D TERMS
BaySaver BaySaver Technologies, Inc.
BFC BayFilter™ Cartridge
BMP Best Management Practice
CWA Clean Water Act
DDC Drain Down Cartridge
DDM Drain Down Module
EMC Event Mean Concentration
FB Field Blank
FD Field Duplicate
FOG Fats, Oil, and Grease
HDPE High Density Polyethylene
LD Laboratory Duplicate
MASWRC Mid-Atlantic Stormwater Research Center
MDE Maryland Department of the Environment
MS4 Large Municipal Stormwater Sewer Systems
NJCAT New Jersey Corporation for Advanced Technology
NJDEP New Jersey Department of Environmental Protection
NPDES National Pollution Discharge Elimination System
PSD Particle Size Distribution
QAPP Quality Assurance Project Plan
QA/QC Quality Assurance/Quality Control
RMHS Richard Montgomery High School
Site Richard Montgomery High School in Rockville, Maryland
SM Standard Method
SSC Suspended Solids Concentration
System BaySeparator™, Horizontal Detention System, and BayFilter™ Vault
TAPE Technological Assessment Protocol Ecology
TARP Technical Assurance Reciprocity Partnership
TMDL Total Maximum Daily Load
TP Total Phosphorus
TSS Total Suspended Solids
USGS United States Geological Survey
UMD University of Maryland
USEPA United States Environmental Protection Agency
1
1.0 I,TRODUCTIO,
Impairment of waterbodies due to stormwater runoff has recently come into focus as an
area of concern by government agencies and municipalities. Stormwater runoff, the dominant
contributor of non-point source pollution, especially in urban areas, carries sediment, nutrients,
and heavy metals into nearby waterbodies. Until recently, much of the government’s and
public’s focus had been on cleaning up the nation’s waterbodies due to point source pollution.
Point source pollution, or the pollution being discharged by an identifiable source, has been a
much easier target for regulations and enforcement by federal and state environmental agencies.
The challenge facing these agencies is regulating the treatment of non-point source pollution,
which does not have a readily identifiable discharge point into a waterbody. Non-point source
pollution is a major source of pollutants in waterbodies nationwide, whether it is from highway
or agriculture runoff.
2.0 GOALS
The evaluation of this field testing program is being done at the request of BaySaver as
part of their submittal for TARP approval of their stormwater treatment devices. The goal of this
evaluation is to audit and evaluate the field testing program currently being conducted by
MASWRC for BaySaver. The scope of work is included in Appendix A. This was accomplished
by examining the existing sampling procedures, laboratory methods, and sample results
reporting, and making recommendations throughout the evaluation process. Precipitation
amounts and flow rates being reported were also analyzed. Concerns identified during the
evaluation are explained in sections 7.6.1 and 7.6.2. Corrective actions taken by MASWRC and
BaySaver to address the concerns raised are discussed in section 7.6.1, and additional
recommendations are in section 7.6.2.
2
3.0 BACKGROU,D
3.1 Federal Regulatory History
The basis for the current regulations governing water quality was enacted in 1948 as the
Federal Water Pollution Control Act. The United States Environmental Protection Agency
(USEPA) did not gain administration and enforcement powers for water quality until 1972, when
the CWA was enacted by the United States Congress. In 1972 the mandate of the CWA was to
reduce industrial (point-source) discharges of pollutants into waterbodies. The National Pollution
Discharge Elimination System (NPDES) was created as part of the CWA. NPDES permits that
governed the discharge of pollution were issued to industry. However, non-point source
pollution, mainly stormwater runoff, was not a subject of these original amendments, and it was
not until 1987 with the Water Quality Acts that non-point source pollution was regulated. The
NPDES permit program was created to regulate large municipal stormwater sewer discharges,
industrial discharges, and discharges into waterbodies that were already impaired. However in
2003, NPDES regulations were modified to include construction sites, 1 to 5 acres in size, and
smaller municipal stormwater sewer discharge in the NPDES permit program. (USEPA, NPDES
History) The EPA has primary oversight of the water quality programs in the United States, but
as part of the CWA, the USEPA has the authority to place responsibility for regulating and
enforcing water quality on the states. Individual states set water quality standards, which are
approved by the USEPA, enforce the standards, and also issue NPDES permits. Maryland was
one of the states granted the authority, by the USEPA, to oversee its own water quality program.
(MDE, Stormwater Program Fact Sheet)
2.2 Local Regulatory Background
In Maryland, the Maryland Department of Environment (MDE) sets, administers, and
enforces the water quality standards for the state of Maryland. If a waterbody is consistently not
reaching the established water quality standard for a given contaminant, it is subject to even
greater regulations and those industries discharging into it have to meet more stringent standards.
A total maximum daily load (TMDL), which is contaminant specific, is assigned to the impaired
waterbody. The TMDL is scientific assessment of the waterbody’s tolerance to that contaminant
and at a load that will not have a detrimental impact on aquatic organisms. The calculation of the
TMDL takes into account both point and non-point sources, plus projected growth and a margin
3
of safety. (MDE, TMDL) The TMDL is used when issuing NPDES permits for discharge into
the impaired waterbody. Once a NPDES permit has been issued to an individual company or
municipality, then it is their responsibility to enact best management practices (BMPs) that meet
the NPDES permit requirement(s). BMPs can be either structural, such as retention ponds, grass
swales, or proprietary treatment devices, such as the BaySeparator™, or non-structural or
housekeeping practices, such as street sweeping the property. It is up to the company or
municipality to balance BMP effectiveness and efficiency with cost. It is this burgeoning market
that BaySaver is concerned with serving through the manufacture and sales of its stormwater
treatment devices.
State approval of stormwater treatment devices is necessary for use in meeting the
NPDES regulations. Because of the variation from state to state regarding requirements for
approval as an acceptable stormwater treatment device, a multi-state coalition, Technology
Acceptance and Reciprocity Partnership (TARP), has been formed, geared towards creating
uniformity and consistency in testing and approval of these devices. TARP is administered by
the New Jersey Corporation for Advanced Technology (NJCAT) a private/public company that
is part of the New Jersey Department of Environmental Protection (NJDEP). TARP approval for
a stormwater treatment device is accepted by the states of California, Illinois, Maryland,
Massachusetts, New Jersey, New York, Pennsylvania, and Virginia. UMD became involved
with the project as part of the Tier II protocol in TARP, and is serving as an independent third
party auditing laboratory analysis and sampling methods, and evaluating the accuracy of the data
reporting.
BaySaver is also applying for approval under the Technology Assessment Protocol-
Ecology (TAPE) program administered by the Washington State Department of Ecology, Water
Quality Program. However, field testing for the TAPE program must be conducted on the west
coast. TAPE has similar requirements to TARP for treatment requirements and what constitutes a
qualified rainfall event. Table 1 shows a comparison of TAPE and TARP requirements.
4
Table 1 - Comparison of TARP and TAPE field testing requirements.
TARP TAPE
Suspended Sediment Analysis SSC and TSS TSS, but SSC = TSS
if sediment <250 microns
Dry Period 6 hour 6 hours
Precipitation Minimum 0.1 inches 0.15 inches
Percent of Rainfall Event
Captured 70% 75%
TSS removal 80% 80%
if TSS between 100 and 200 mg/L
effluent TSS <20 mg/L
if TSS <100 mg/L
>80%
if TSS >200 mg/L
50%
if TSS between 100 and 200 mg/L
(pretreatment)
effluent TSS <50 mg/L
if TSS <100 mg/L (pretreatment)
Rainfall Event Duration none 1 hour
Field Duplicates required 10% of samples
2.3 Stormwater Treatment
Urban areas are largely impervious surfaces, such as asphalt or concrete pavement or
roofs; various pollutants are picked up by the flow of stormwater across them. Dry deposition of
contaminants on impervious surfaces creates an accumulation on the surface, until a rainfall
event mobilizes the contaminants (sediment, heavy metals, etc.). Excessive stormwater runoff
can inundate surface waterbodies with pollutants and nutrients leached from the ground surface.
The excess nutrients (phosphorus, nitrogen) can cause algal blooms and increase biological
growth in the water, which ultimately depletes the dissolved oxygen (eutrophication) thus
inhibiting growth of other organisms (fish). Land development often reduces the amount and
diversity of vegetation on the land. The decrease in vegetation limits the amount of water trapped
during a precipitation event. Land development also adds more impervious surfaces, which
increases stormwater runoff by decreasing the land available for infiltration.
5
Stormwater sewers that carry stormwater runoff into a constructed wetland or
sedimentation basin, prior to discharge to a surface water body, would allow for suspended
sediments to settle out and nutrient uptake by aquatic microorganisms and vegetation. However,
the area required for a basin or wetland is often not available or is too valuable. This is
especially relevant in urban settings, where the cost per acre of open land makes it financially
unattractive to use as a basin. Instead, a stormwater treatment device (such as the
BaySeparator™) can be installed underground in a parking lot. Spatial and temporal constraints
are addressed through the use of proprietary stormwater treatment devices as structural BMPs.
BaySaver Technologies, Inc. headquartered in Mount Airy, MD is in the business of
developing and manufacturing stormwater treatment devices that address the problems
associated with stormwater runoff. Their current line of products includes the BaySeparator™
and BayFilter™, both which were tested under field conditions as part of this project.
6
3.0 THE TREATME,T SYSTEM
3.1 The Site
The site is located at Richard Montgomery High School in Rockville, Maryland, Figure
1. The system treats stormwater runoff from an approximately 3.6 acre drainage area, which is
approximately 83% impervious surface, mainly asphalt pavement, concrete, and the building
roof, and 17% landscaped or grassy areas (medians in the parking lot), Figure 2.
Figure 1 - As-built drawing of the site (Courtesy of BaySaver). Arrow marks the approximate
location of the treatment devices.
7
Figure 2 - View of Richard Montgomery High School parking lot (the site). Arrow
on the picture marks the approximate location of the treatment devices.
3.2 The System
The system consists of a BaySeparator™ model 3K unit (Figure 3) and a pre-cast 8 by 12
foot concrete vault (BayFilter™ vault) containing 5 BayFilter™ Cartridges (BFC) and 1 drain
down cartridge (DDC). The system was installed in March 2008 as part of construction activities
occurring at the site. The configuration was altered in June 2008, according to the MASWRC,
when the DDC was removed and a drain down module (DDM) was installed with each BFC. The
stormwater runoff is collected by inlets in the parking lot and flow is carried to the
BaySeparator™, Figure 4. Flow from the BaySeparator™ goes into a diversion structure, a pre-
cast concrete manhole, then into a horizontal detention system. The horizontal detention system
is constructed of three 8-foot diameter corrugated metal pipes, a total of 205 feet in length.
8
Figure 3 - Overview of the BaySeparator (Courtesy of BaySaver).
Figure 4 - Layout of the BaySeparator™ and BayFilter™ (Courtesy of BaySaver).
9
For the purposes of this study the BayFilter™ system consists of both the horizontal
detention system and the filter vault. The stormwater treatment performance of the filter vault
can not be properly evaluated separate from the horizontal detention system, because flow paced
samples were taken at the outlet from the BaySeparator™, in the diversion structure. The outlet
of the BaySeparator™ is considered the same as the inlet to the BayFilter™ system. Ultimately,
treated stormwater from the filter vault is discharged through 4-inch PVC underdrains to two
stormwater management ponds located on the Richard Montgomery High School property. The
horizontal detentions system, 10,300 cubic foot volume, is designed to store stormwater runoff
for a 1-inch rainfall event on a 3.6 acre drainage area. The horizontal detention system has a dry
storage of approximately 6 inches in the bottom of the pipes.
3.3 Site Characterization
The site was evaluated in March 2008, shortly after the BaySeparator™ was installed, for
suitability as a site to conduct field testing under the NJDEP Tier II protocol. The Tier II protocol
calls for collecting water quality samples to determine influent suspended solids concentration
and particle size distribution (PSD). To qualify, the site needs to have stormwater runoff with
TSS concentrations between 100 to 300 mg/L and contain sediment particles with a diameter less
than 100 µm.
Five rainfall events, March 8, March 16, March 20, April 1, and April 6, 2008 had TSS
influent concentrations of 180, 29, 64, 9, and 33 mg/L respectively. Only one of the rainfall
events resulted in an influent TSS concentration within the Tier II protocol range of 100 to 300
mg/L. PSD analysis was conducted on samples collected from the March 8, 2008 rainfall event.
Results indicated that 74% of the influent particles had a diameter less than 100 µm. The PSD
analysis was conducted by the Particle Engineering Research Center at the University of Florida.
10
4.0 BAYSAVER STORMWATER TREATME,T DEVICES
4.1 The BaySeparator™
The BaySeparator™ is a hydrodynamic separator that uses gravity to settle out sediment.
According to the BaySeparator™ Technical and Design Manual, 80% of suspended sediments
are removed by the BaySeparator™. The BaySeparator™ consists of a primary storage chamber
(primary manhole), an internal HDPE separator (BaySaver) and secondary storage chamber
(storage manhole), Figures 3, 5, and 6. The stormwater flows from the parking lot into the inlet
pipe, then into the primary manhole. Large sediment particles settle out in the primary manhole.
The flow continues from the primary into the storage manhole, where the fine sediments settle
out.
Figure 5 - Plan View of the BaySeparator™ (Courtesy of BaySaver)
11
Fats, oils, and grease (FOG) are collected in the secondary manhole as floatable materials
(Figures 5 and 7). The treatment capacity of the BaySeparator™ model 3K unit is 3.4 cfs or 25.4
gpm. Above 25.4 gpm, the flow bypasses the secondary manhole and is discharged through the
outlet pipe into the horizontal detention system with little or no treatment. The internal flow
splitter (Figures 6 and 8) governs the three flow paths that stormwater can take: 1. flow from the
primary to secondary manhole; 2. flow from the secondary manhole to the outlet; and 3. flow
bypass to the outlet.
Figure 6 - Profile View of the BaySeparator™ Primary Manhole (Courtesy of
BaySaver).
12
Figure 7 - View of trash collected in the storage manhole.
Figure 8 - View of the internal flow splitter in the primary manhole.
If the horizontal detention system is full, and the runoff flow rate is greater than the
BayFilter™ flow rate, then the excess flow is discharged from the diversion structure directly
into the stormwater management pond, Figure 4.
13
4.2 The BayFilter™
The BayFilter™ is a media filter used to remove fine suspended sediments, organics,
nutrients, and heavy metals from stormwater. The historical use of sand filters in water and
wastewater treatment is the basis for the BayFilter™. The media and filter fabric trap suspended
contaminants and the contaminants are adsorbed onto the surface of the media particles. The
adsorption is driven by the high surface area of the media particles combined with the ionic
interactions between the media particles and the contaminants. The BayFilter™ cartridge (BFC)
consists of a filter fabric with layers of media, Figure 9. The media used in the BFC is a mix of
sand, perlite, and activated alumina designed for maximum nutrient, heavy metal, and organics
removal. There are 43 square feet of surface area in each BFC that filters stormwater. The
filtering process is gravity driven.
When the level in the filter vault reaches 28 inches, the height of the BFC, the stormwater
flows through the media and the treated stormwater is forced up to the top of the BFC. The
treated stormwater flows out through the center of the BFC into the outlet pipe. A siphon allows
for treatment of stormwater until the water level in the filter vault reaches 6 inches, the bottom of
the BFC. Once the water drops below 6 inches (bottom of the BFC), the siphon is broken and the
filter is backwashed with the treated stormwater remaining in the BFC. This backwashing
process re-suspends sediment trapped by the filter and flows out into the filter vault, where
sediment settles out. The drain down cartridge (DDC) was used to completely drain the filter
vault, because the BFC is located 6 inches from the bottom of the vault. However, in June 2008
the DDC was removed and 5 DDMs, one for each BFC, were installed. The DDM (Figure 10) is
a 4-inch PVC pipe with a sand filter inside and is connected to the underdrains. The DDM
functions similar to the DDC in treating the remaining 6 inches of stormwater in the vault. The
filter vault at the site contains 5 BFCs as depicted in Figure 11.
14
Figure 9 - Profile view of the BayFilter™ cartridge (Courtesy of BaySaver).
Figure 10 - View of the BayFilter™ draw down module (Courtesy of BaySaver).
15
Figure 11 - Plan view of the BayFilter™ vault at the RMHS site (Courtesy of BaySaver).
The horizontal detention system is an integral part of the overall treatment system at the
Richard Montgomery site; it attenuates the flow rate of the stormwater. The attenuation is
necessary, because the number of BFCs used is not capable of treating the stormwater at the
influent flow rate. The filter vault is cleaned of sediment and the BFCs exchanged when the filter
vault does not fully drain within forty hours of the end of the stormwater inflow.
16
5.0 SAMPLI,G
5.1 TARP Tier II Protocol Requirements
An event is considered qualified if: there has been at least a 6 hour dry period prior to
subsequent rainfall; there is greater than 0.1 inches of precipitation; and at least 70% of the
stormwater runoff volume has been sampled. It should be noted that BaySaver is considering a
qualified event to have greater than 0.15 inches of precipitation, which is the requirement for
TAPE. The Tier II protocol, Appendix B, requires at least 15 qualified events and recommends at
least 20 qualified events. In addition to the minimum number of qualified events, the total
precipitation from the qualified events must be at least 50% of the total annual rainfall. The site
has an annual rainfall of approximately 42 inches (NOAA, Climate of Maryland), therefore, the
qualified events need to have a cumulative precipitation of greater than 21 inches. As of
November 14, 2008, a total of 17 qualified events have been sampled with a total of 10.3 inches
of precipitation. Tier II protocol requires some of the qualified events to be “adverse conditions”
for the stormwater treatment device. Of the current qualified events, at least 2 events can be
considered adverse conditions, which have high intensity or long duration rainfall events that
may approach or exceed the design capacity of the system. Requirements relating to site
qualification, sample collection, and laboratory analysis are discussed as part of each specific
section (TARP Protocol, section 3.3.1.3).
5.2 QAPP
A Quality Assurance Project Plan (QAPP) was developed by BaySaver to provide
guidance to MASWRC when conducting the field testing of the system, Appendix C. The QAPP
is site specific and is meant to ensure that sampling and analysis of field data is done safely and
accurately. According to the QAPP, MASWRC will conduct quality assurance/quality control
(QA/QC) on the water quality samples throughout the duration of the field testing program. The
QA/QC is a required part of the TARP Tier II protocol, and is necessary to ensure accuracy of
data reporting. The QAPP was prepared and edited by BaySaver during the course of the
evaluation process. A final version was submitted to NJCAT on November 5, 2008.
17
5.3 Qualified Events
A qualified event must meet certain criteria outlined in the TARP protocol. The rainfall
event has to have greater than 0.1 inches of precipitation (TARP Protocol section 3.3.1.2), and
the QAPP identifies 0.15 inches as the minimum qualified rainfall event being used (BaySaver,
QAPP). The water quality samples collected must be representative samples that cover 70% of
the stormwater runoff volume during an event (TARP Protocol, section 3.3.1.2). In addition, a
minimum of 10 influent and 10 effluent samples need to be collected (TARP Protocol, section
3.3.1.2).
5.4 Experimental Methods
The sampling and analytical methods discussed below include both the methods used and
procedures conducted by MASWRC in gathering data. Where applicable, the analytical methods
used by UMD in determining the concentrations of select water quality constituents are
identified below.
5.4.1 Field Sampling
The site is equipped with Rainwise rain gauge and HOBO logger to gather precipitation
data. The rain gauge collects and records precipitation in 0.01 inch increments using a tipping
bucket method. The rain gauge and logger are installed on an 8-foot post located approximately
20 feet from the BaySeparator™.
The flow weighted water quality samples are collected using three ISCO 3700 auto-
samplers that are connected to ISCO 4250 area/velocity flow meters. One flow meter is located
in the inlet pipe to the primary manhole in the BaySeparator and it triggers the two auto-samplers
in the BaySeparator™. The auto-samplers in the BaySeparator™ collect samples at the inlet to
the primary manhole and the outlet from the secondary manhole, Figure 12. These sample
locations are termed BaySeparator™ Influent and BaySeparator™ Effluent. The second flow
meter is located and the outlet from the filter vault, and governs the flow pacing of the auto-
sampler located there. This sample location is termed BayFilter™ Effluent. (Figure 4)
18
Figure 12 - View of the BaySeparator™ influent flow meter and effluent auto-sampler set-up.
The auto-samplers are flow paced, collecting a discrete sample (200 mL to 250 mL), for a
set volume of flow (2000 to 6000 L). Each sample bottle contains a composite of 4 discrete
samples for a total of 0.8 to 1 liter. The samples are taken using a 3/8 inch suction line, which is
set 0.5 inches off the bottom of the pipe, Figure 13. The volume of pacing is based upon
expected intensity of the rainfall event. The auto-samplers are programmed to purge and rinse
the suction line between samples, reducing the potential for cross contamination between
samples.
19
Figure 13 - View of the inlet pipe to the BaySeparator™, including the suction line
and flow meter sensor.
The flow meters and auto-samplers were calibrated by MASWRC prior to installation.
There is no set schedule to replace flow meters and check calibration. Currently, MASWRC will
replace a flow meter if the readout in the vault does not properly balance after an event.
MASWRC visually inspects and checks programmed settings after every rainfall event, even if
the flow is less than 0.1 inches, a non-qualifying event. The deep cycle marine batteries used to
power the instruments are checked weekly and exchanged regularly.
The samples are collected by MASWRC within 24 hours of the end of the rainfall event
and are transported to the MASWRC laboratory for analysis of water quality parameters. (Figure
13) The sample bottles that are collected by MASWRC are replaced with clean sample bottles
after each event. The sample bottles are washed and prepared at the MASWRC laboratory with
an Alconox solution (non-phosphorus containing detergent) and rinsed with deionized water.
20
Figure 13 - View of cooler used to transport sample bottles. The colored labels (3
sheets) next to the cooler are used to color code the bottles based on the location of
the auto-sampler (BaySeparator™ influent and effluent and BayFilter™ effluent).
5.4.2 UMD Sample Collection
For the October 25 and November 13, 2008 rainfall events, selected samples were split in
the field by re-suspending the sample in the sample bottle by rapid shaking. The top half of the
sample was poured into another sample bottle, which was taken back to the UMD lab for water
quality analysis.
5.4.3 Sampling Protocol
Field blanks are specified in the QAPP to be analyzed twice during the field testing
program, before the first stormwater runoff event and the mid-point of the program (between
events 6 and 9). All three auto-samplers should have field blanks collected using deionized
water. The field blanks are used to quantify any effects the equipment might have on the water
quality samples. To date, field blanks were collected from the BaySeparator™ influent and
effluent auto-samplers. The field blank results were less than 1 mg/L for SSC and less than 0.1
mg/L for TP. These results are at or below the detection limit of the analysis methods. A second
set of field blanks has not been collected yet.
21
Field duplicates are to be collected and analyzed over the course of the program at least
three times. The water quality samples collected for analysis, including SSC, TP, and turbidity,
should be split and the two results compared for representativeness. The samples were split for
SSC, starting with the May 31, 2008 event. For the April 11, 2008 event, duplicate samples were
collected and analyzed for TP. In a sample to duplicate sample comparison, the results have a
range in percentage difference from 0 to 100%. However, the mean values (sample and duplicate
samples) for the BaySeparator™ influent (46 and 47 mg/L), effluent (34 and 34 mg/L), and
BayFilter™ effluent samples (0.27 and 0.23 mg/L) show no significant difference. While there
may be some variability on a sample to sample basis, the overall agreement between results is
satisfactory. However, no other water quality constituents have had duplicates collected or
analyzed.
Tier II protocol and the QAPP both specify a chain of custody for all water quality
samples collected. The chain of custody is used to track samples and protect sample integrity.
Prior to the end of September 2008, MASWRC was not using a formal chain of custody.
However, samples were being collected from only one site for the duration of the field testing
program and MASWRC personnel retained custody of the samples upon collection at the site. A
copy of the chain of custody currently being used is included in Appendix D.
A key component of any field testing program, and a requirement of the Tier II protocol
is a health and safety plan. As part of the health and safety component of the QAPP and Tier II
protocol, confined space entry training and permit is required. Not only is this a requirement of
the documents governing the field testing program, but also the Occupational Safety and Health
Administration (OSHA) regulations. The appropriate OSHA regulations can be found in 29 CFR
1910.146 (Permit-required confined space). Currently, no health and safety plan has been
developed. However, confined space entry training was conducted for MASWRC personnel.
5.4.4 Analytical Methods
The samples are collected in the field in plastic sample bottles, then transported back to
the MASWRC laboratory, where they are analyzed. Although other water quality constituents
are analyzed by MASWRC, under the scope of this project, UMD was only concerned with
suspend sediment concentration (SSC), total phosphorus (TP), and turbidity.
22
5.4.4.1 Suspended Sediment
MASWRC uses TSS and SSC to determine the concentration of suspended sediment in a
sample. There has been an ongoing discussion in the scientific community regarding the proper
method for determining concentrations of suspended sediment in samples. The historical method
(water and wastewater industry) has been total suspended solids (TSS), where a representative
100 mL sample is withdrawn from the center of a well mixed sample. There has been a trend to
analyze stormwater samples using SSC, where the entire sample volume is used.
The United States Geological Survey (USGS) has been leading the push towards using
SSC to analyze stormwater, because it is the most representative indication of sediment in a
sample. The USGS argues that SSC is better for natural systems, and that TSS is being
misapplied. They have found that there is no correlation between TSS and SSC because of the
wide range of conditions possible in natural systems (Gray et al., 2000). In environments where
larger particles are more prevalent (i.e., urban environments with sand), the TSS concentrations
have been less than SSC, because the larger particles are not being sampled from the center of
the container. Other studies have found a closer correlation between TSS and SSC, but still
regard SSC as the more accurate representation of the entire sample (Clark and Pitt, 2008).
5.4.4.1.1 MASWRC Analyses
Sampling and analysis for SSC is done according to the American Society for Testing
and Materials (ASTM) D 3977-97 standard. The glass fiber filter (Whatman A/H 90 mm) is
prepped by filtering approximately 30 mL of deionized water through it, then drying it for at
least an hour at 103oC to 105
oC. The filters are stored in aluminum weighing pans in a humidity
controlled container prior to weighing and use.
A stormwater sample is weighed in the sample bottle, then poured into a glass jar with a
stirrer bar and the sediment is re-suspended. A 60 mL sub-sample is withdrawn by pipette for
turbidity, color, and total phosphorus analysis. The empty sample bottle is weighed, the
difference between initial full mass and the empty bottle mass, minus the sub-sample, is
calculated as the suspended sediment sample volume (assuming a density of 1 g/mL).
Half the sample in the glass jar is poured back into the sample bottle, weighed, and
poured through the filter using a vacuum pump. The remaining sample in the glass jar is poured
through a second filter, and the glass jar and sample bottle are rinsed with deionized water into
the filter. The filters are placed in the drying oven (103oC to 105
oC) for at least one hour, and
23
then are weighed. The difference in filter mass determines the mass of sediment in each part of
the sample. The SSC for each of the two parts of the sample are calculated and recorded as
sample A and B. The mean SSC value is based on the total sediment mass and the total sample
volume filtered. Samples collected from high intensity rainfall events are sieved for large
particles (greater than 250 µm). Sieving has occurred, for the April 20, 2008 rainfall event, a
non-qualifying event, and the August 14 and 28, 2008 rainfall events.
Since the June 3, 2008 rainfall event, MASWRC has been splitting the samples as
describe above. Prior to that, the whole sample volume was filtered through one filter. By
splitting the sample into an A and B part, MASWRC uses these as two TSS samples. The current
SM 2540 uses a 100 mL sub-sample for TSS, and previous standard methods require the sub-
sample to be collected with a wide-bore pipette (Clark and Siu, 2008).
It should be noted that SM 2540 requires samples to be preserved at 4oC to limit bacterial
decomposition. Samples should never be held more than 7 days, and if possible, analyzed in less
than 24 hours after collection. Since the end of September 2008, MASWRC has been using ice to
transport samples from the site and a refrigerator to store samples in the laboratory prior to
analysis. Prior to that event, samples were not kept below 4oC, but often were analyzed
immediately following collection. There were times when multiple rainfall events in close
succession prevented the immediate analysis of samples.
5.4.4.1.2 UMD Analyses
SSC was determined using ASTM D 3977-97. The glass fiber filters (VWR 90 mm) were
prepared the same way as MASWRC. The sample was re-suspended by hand (shaking) and a
sub-sample (approximately 100 mL) was taken from the sample bottle for TP and turbidity
analyses. The sample bottle was weighed with the sample in it and at the end as an empty sample
bottle to determine volume filtered. Because the sample was a split sample, it was not necessary
to weigh an “A” and “B” sample as MASWRC does; only one sample was filtered. The filters
were dried for at least 24 hours at 103oC to 105
oC, then weighed to determine sediment mass.
The SSC was calculated using a density of 1 g/mL for the sample mass, and the difference in
mass between the prepped filter and the filter with sediment.
24
5.3.4.2 Total Phosphorus Analysis
Total phosphorus concentration is determined using persulfate digestion to convert
organic phosphate into orthophosphate (PO4) and measured using the ascorbic acid method
(colorimetric measurement). Persulfate digestion is conducted according to the SM 4500 method,
by MASWRC and UMD. The only difference is that MASWRC uses a 20 mL sub-sample,
instead of the 30 mL sub-sample specified by SM 4500.
MASWRC uses the ascorbic acid molybdenum blue method (EPA 365.2) with an
ascorbic acid reagent from HACH to determine the total phosphorus (TP) concentration.
Samples were analyzed by MASWRC using a HACH DR-2800 spectrophotometer. A 1.5
dilution factor is used in calculating TP concentrations, because the original sample was 20 mL,
but the current 30 mL sample volume was used to determine the colorimetric measurement (20
mL original volume:30 mL current volume).
UMD analyzed the samples according to the ascorbic acid method in SM 4500. UMD
correlated spectrophotometer results to known TP standards (range 0 to 1.0 mg/L). For both
UMD and MASWRC, the “A” and “B” samples are the same sample measured with the
spectrometer twice. A mean value of the “A” and “B” samples is reported as the samples value.
5.3.4.3 Turbidity
Turbidity is analyzed by both UMD and MASWRC according to SM 2130
(Nephelometric method). The “A” sample was the first reading, then withdrawn, wiped clean
again, and re-run to determine the “B” sample. Occasionally, turbidity samples were not
immediately analyzed by MASWRC following sample collection. Since the August 2008 events,
MASWRC has been running turbidity samples immediately upon sample collection.
25
6.0 RESULTS
6.1 Precipitation
Precipitation data gathered at the site by MASWRC was compared to three other sites in
the surrounding area (Rockville, MD). Precipitation data were obtained from
www.wunderground.com (zip code 20852) for each of the rainfall events monitored by
MASWRC. Not all the dates had data available for them on www.wunderground.com. The
rainfall event number corresponds to the date samples were collected (Table 2). The data
collected at the site (RMHS) was within the range of the surrounding sites, Figure 14.
Table 2 - Correlation of date, rainfall event number, precipitation amount
measured, and percentage of total precipitation.
Event Date Rainfall
Event Precipitation (in)
Percentage of
Total
Precipitation
11-Apr-08 #1 0.29 3%
26-Apr-08 #2 0.58 6%
28-Apr-08 #3 0.63 6%
31-May-08 #4 0.26 3%
3-Jun-08 #5 0.55 5%
4-Jun-08 #6 0.62 6%
16-Jun-08 #7 0.40 4%
23-Jun-08 #8 0.24 2%
27-Jun-08 #9 0.72 7%
9-Jul-08 #10 0.44 4%
13-Jul-08 #11 0.94 9%
23-Jul-08 #12 0.46 4%
2-Aug-08 #13 0.26 3%
14-Aug-08 #14 0.28 3%
28-Aug-08 #15 1.75 17%
25-Oct-08 #16 1.16 11%
13-Nov-08 #17 0.75 7%
Total -- 10.33 100%
26
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
#1 #2 #3 #4 #5 and
#6
#7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17
Rainfall Event
Precipitation (in)
Site
Woodley Gardens
Potomac Highlands
Flower Valley
Figure 14 - Graph of precipitation from the site (RMHS) and three surrounding locations. Surrounding
locations precipitation data was obtained from www.wunderground.com.
6.2 Runoff/Volume Balance
Precipitation data gathered from the site by MASWRC was used to determine the
precipitation volume, Table 3. The site is 3.62 acres and a volume (gal) was calculated using the
rainfall depth. The BaySeparator™ and BayFilter™ flow volumes (gal) were calculated from the
ISCO flow meter data, which directly measured cumulative volumes (liters). The runoff percent
is the BaySeparator™ flow volume divided by the precipitation volume. The runoff percent was
on average 60% of the rainfall measured by the rain gauge, which is to be expected on a site
largely covered by an impervious surface, Table 3. The site was approximately 83% impervious
surfaces. The rational method runoff coefficients for this site, based on 83% impervious and 17%
pervious landscaping, would be between 0.61 and 0.82. The average runoff coefficient measured
on the site (0.6) is on the low side of that range, but still acceptable. Based upon reasonable
agreement between the BaySeparator™ and precipitation volume, some questions arrise
regarding the BayFilter™ volume, Figure 15. The flow into the BaySeparator™ should be the
same volume as the flow into the BayFilter™. The unaccounted flow (fraction not going to the
27
BayFilter™) is the BayFilter™ flow volume divided by the BaySeparator™ flow volume. On
average there was 24% less flow measured into the BayFilter™ as into the BaySeparator™,
Table 3.
Table 3 - Precipitation volume calculated from rainfall data collected and BaySeparator™ and BayFilter™
volumes recorded by the flow meters. Average Runoff and Percent ,ot Going to BayFilter™ are flow
weighted averages based on precipitation for each event.
Event Precipitation
Volume (gal)
BaySeparator™
Flow Volume
(gal)
Runoff
BayFilter™
Flow
Volume
(gal)
BaySeparator™
and BayFilter™
Difference (gal)
Percent ,ot
Going to
BayFilter™
11-Apr-08 28,000 15,000 53% 8,000 7,000 47%
26-Apr-08 55,000 32,000 58% 25,000 7,000 21%
28-Apr-08 60,000 39,000 64% 34,000 5,000 13%
31-May-08 25,000 17,000 67% 11,000 6,000 35%
3-Jun-08 52,000 30,000 58% 22,000 8,000 27%
4-Jun-08 59,000 36,000 61% 24,000 12,000 34%
16-Jun-08 38,000 23,000 60% 17,000 6,000 25%
23-Jun-08 23,000 13,000 55% 11,000 2,000 16%
27-Jun-08 68,000 45,000 65% 39,000 5,000 12%
9-Jul-08 42,000 23,000 56% 22,000 1,000 4%
13-Jul-08 89,000 52,000 58% 41,000 11,000 21%
23-Jul-08 44,000 28,000 64% 20,000 7,000 27%
2-Aug-08 25,000 16,000 64% 11,000 4,000 28%
14-Aug-08 27,000 13,000 50% 7,000 6,000 45%
28-Aug-08 166,000 104,000 62% 73,000 31,000 30%
25-Oct-08 110,000 60,000 55% 55,000 6,000 9%
13-Nov-08 71,000 41,000 57% 23,000 18,000 44%
Average 80,000 48,000 60% 36,000 12,000 24%
Total 982,000 587,000 -- 443,000 142,000 --
28
000
20,000
40,000
60,000
80,000
100,000
120,000
140,000
160,000
180,000
#1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17
Rainfall Event
Volume (gal)
Precipitation
BaySeparator™
BayFilter™
Figure 15 - Graph of the precipitation, BaySeparator™, and BayFilter™ volumes by rainfall event.
6.3 Field Flow Test
Due to the differences in flow volumes between the BaySeparator™ and BayFilter™, a
field flow test was conducted on November 9, 2008. A water truck, filled with potable water,
was used to provide a measurable steady flow rate, Figure 16. The BayFilter™ effluent flow
meter was measured first. A 2-inch flexible hose was attached to the underdrain system in the
filter vault, Figure 17. The water truck had a 2-inch flexible hose connected to straight section of
approximately 20 feet of rigid PVC pipe. The flow rate was governed by a PVC ball valve at the
beginning of the rigid PVC pipe. Two electronic flow meters were inline with the rigid section of
pipe. The aboveground flow meters were factory calibrated to within 3% according to the
manufacturer. The ball valve was adjusted to reach 7 discrete steady flow rates. The flow rate
being measured in the filter vault was allowed to reach a steady state and the corresponding flow
rate on the aboveground meters was recorded. Table 4 shows the recorded flow rates and the
difference (gpm) between the aboveground flow meters and the ISCO flow meter in the filter
vault.
29
Figure 16 - View of the 2-inch flex hose connected to the BayFilter™ underdrain
system (,ovember 9, 2008).
Figure 17 - Water truck used during field flow test. Two flow meters are located on
the PVC piping in the picture (,ovember 9, 2008).
30
Table 4 – BayFilter™ flow rates from field flow test conducted on
,ovember 9, 2008.
BayFilter™ Vault
Aboveground Flow Meters BayFilter
#1
(gpm)
#2
(gpm)
Mean
Flow
(gpm)
ISCO
Flow
Meter
(gpm)
Difference
(gpm) Difference
22 23 22 13 9 41%
33 35 34 22 12 35%
43 46 44 33 12 27%
56 61 59 44 15 26%
67 71 69 56 13 19%
76 80 78 67 11 14%
The BaySeparator™ flow meter was measured second. The same hose was run from the
water truck to the rigid section of pipe, but only one flow meter was used, because of good
agreement between the flow meters. A second 2-inch flexible hose was run from the rigid section
to the inlet closest to the BaySeparator™, Figure 18. The water in the flow test entered the
concrete inlet pipe directly, and was approximately 30 feet from the flow meter in the
BaySeparator™. The same procedure as before was used; flow rates reached a steady state and
an average of the steady state readings was taken. The results of the flow rate test in the
BaySeparator™ are listed in Table 5. The correlated rainfall intensity was calculated by the
Rational Method using the drainage area of 3.62 acres and a runoff coefficient of 0.6 and the
aboveground flow meter data.
31
Figure 18 – Flow test hose entering the inlet closest to the BaySeparator
(approximate location marked by the arrow). (,ovember 9, 2008)
Table 5 – BaySeparator™ flow rates from field flow test
conducted on ,ovember 9, 2008.
BaySeparator™
Flow
(gpm)
ISCO
Flow
Meter
(gpm)
Difference
(gpm) Difference
Correlated
Rainfall
Intensity
(in/hr)
22 25 -3 12% 0.02
41 43 -2 4% 0.04
72 68 4 -5% 0.07
95 175 -80 84% 0.10
110 188 -78 70% 0.11
147 285 -138 94% 0.15
6.3 Monitoring Results
The results being reported as part of this testing program have been submitted to NJCAT
twice, in June 2008 and November 2008. The results reported are summarized below in two
sections, Constituent Concentrations (Event Mean Concentrations) and Removal Efficiency
Values.
32
6.3.1 Constituent Concentrations (Event Mean Concentrations)
The results in Tables 6, 7, and 8 show the calculated values and the reported event mean
concentrations (EMC). The calculated values are the event mean concentrations calculated by
UMD from the raw data supplied by MASWRC. In the case of SSC, the solids mass and sample
volumes were calculated for each sample taken, and an event mean concentration was
determined. In instances where a sample was not collected (i.e., an empty sample bottle), the
mean value of the samples collected before and after that missing sample was calculated and
used for that missing sample value (i.e., sample before the missing sample with a SSC
concentration of 100 mg/L and a sample after the missing sample with a SSC concentration of 50
mg/L would result in a calculated SSC of 75 mg/L for the missing sample value). There were
five events that had an uncollected sample.
In some instances, only half the expected sample volume was collected in the sample
bottle; this occurred in 8 events. In those instances, splitting the sample into “A” and “B”
samples for SSC analysis was not possible. The entire volume was analyzed as one sample and
the result calculated the same as the mean values determined from the “A” and “B” samples for
the other sample bottles.
The reported values are those event mean concentrations reported by MASWRC to
BaySaver in their September 19, 2008 report (Liu, 2008). Reported values highlighted in red are
values that are much greater or smaller than the calculated values (+ 5%). It should be noted that
the reported values for TP from the May 31, 2008 and June 3, 2008 event appear to be flipped.
33
Table 6 – UMD calculated and reported EMC values for SSC. Reported values were reported by MASWRC
to BaySaver on September 19, 2008. Reported values in red signify an EMC 5% less than or greater than the
EMC calculated by UMD. Average calculated EMC value is flow weighted based on event precipitation.
BaySeparator™ Influent BaySeparator™ Effluent BayFilter™ Effluent
Event Calculated
SSC (mg/L)
Reported
SSC (mg/L)
Calculated
SSC (mg/L)
Reported
SSC
(mg/L)
Calculated
SSC (mg/L)
Reported
SSC (mg/L)
11-Apr-08 718 718 166 170 19 19
26-Apr-08 129 129 65 65 9 9
28-Apr-08 47 47 20 20 6 6
31-May-08 168 168 104 104 12 12
3-Jun-08 185 185 97 98 6 6
4-Jun-08 184 184 105 105 13 13
16-Jun-08 28 28 38 38 8 8
23-Jun-08 47 47 50 50 13 13
27-Jun-08 225 225 131 138 17 18
9-Jul-08 350 350 201 201 23 23
13-Jul-08 105 100 42 40 18 17
23-Jul-08 81 81 45 45 17 17
2-Aug-08 480 412 181 174 20 20
14-Aug-08 636 642 272 286 25 26
Average
(Prior to 28-Aug-
08)
202 -- 96 -- 14 --
28-Aug-08 150 73 11
25-Oct-08 237 185 19
13-Nov-08 24
Not Reported
Yet
34
Not
Reported
Yet 12
Not Reported
Yet
Overall 184 -- 97 -- 14 --
34
Table 7 – UMD calculated and reported EMC values for TP. Reported values were reported by MASWRC to
BaySaver on September 19, 2008. Reported values in red signify an EMC 5% less than or greater than the
EMC calculated by UMD. Average calculated EMC value is flow weighted based on event precipitation.
BaySeparator™ Influent BaySeparator™ Effluent BayFilter™ Effluent
Event Calculated
TP (mg/L)
Reported TP
(mg/L)
Calculated
TP (mg/L)
Reported
TP (mg/L)
Calculated
TP (mg/L)
Reported
TP (mg/L)
11-Apr-08 0.46 0.46 0.34 0.34 0.27 0.27
26-Apr-08 0.47 0.47 0.44 0.44 0.01 0.01
28-Apr-08 0.16 0.16 0.08 0.08 0.04 0.04
31-May-08 0.41 0.29 0.35 0.27 0.26 0.12
3-Jun-08 0.29 0.41 0.27 0.35 0.12 0.26
4-Jun-08 0.56 0.56 0.25 0.25 0.11 0.11
16-Jun-08 0.16 0.16 0.18 0.18 0.07 0.07
23-Jun-08 0.30 0.30 0.23 0.23 0.11 0.11
27-Jun-08 0.42 0.42 0.41 0.41 0.07 0.07
9-Jul-08 0.45 0.45 0.46 0.46 0.14 0.14
13-Jul-08 0.49 0.49 0.18 0.18 0.09 0.06
23-Jul-08 0.28 0.28 0.15 0.15 0.07 0.07
2-Aug-08 1.56 1.35 0.71 0.63 0.16 0.13
14-Aug-08 0.82 0.82 0.54 0.55 0.30 0.30
Average
(Prior to 28-Aug-08) 0.45 -- 0.30 -- 0.11 --
28-Aug-08 0.29 0.37 0.16
25-Oct-08 0.34 0.35 0.21
13-Nov-08 0.15
Not Reported
Yet
0.22
Not Reported
Yet
0.14
Not Reported
Yet
Overall 0.39 -- 0.31 -- 0.13 --
35
Table 8 – UMD calculated and reported EMC values for turbidity. Reported values were reported by
MASWRC to BaySaver on September 19, 2008. Reported values in red signify an EMC 5% less than or
greater than the EMC calculated by UMD. Average calculated EMC value is flow weighted based on event
precipitation.
BaySeparator™ Influent BaySeparator™ Effluent BayFilter™ Effluent
Event
Calculated
Turbidity
(,TU)
Reported
Turbidity
(,TU)
Calculated
Turbidity
(,TU)
Reported
Turbidity
(,TU)
Calculated
Turbidity
(,TU)
Reported
Turbidity
(,TU)
11-Apr-08 44 44 44 44 13 13
26-Apr-08 23 23 26 26 5 5
28-Apr-08 9 9 13 13 7 7
31-May-08 37 37 38 38 10 10
3-Jun-08 29 29 27 27 9 9
4-Jun-08 101 70 62 62 9 9
16-Jun-08 16 16 16 16 10 10
23-Jun-08 21 21 23 23 10 10
27-Jun-08 66 66 70 70 23 23
9-Jul-08 44 44 51 51 15 18
13-Jul-08 23 24 21 21 13 13
23-Jul-08 29 29 28 28 19 19
2-Aug-08 144 126 80 78 16 16
14-Aug-08 162 162 101 101 28 28
Average
(Prior to 28-Aug-08) 47 -- 40 -- 13 --
28-Aug-08 33 34 11
25-Oct-08 164 148 57
13-Nov-08 13
Not Reported
Yet
17
Not
Reported
Yet 12
Not Reported
Yet
Overall 56 -- 49 -- 18 --
6.3.2 Removal Efficiency
The removal efficiencies of the BaySeparator™, BayFilter™, and the system were
presented in the same reports as the EMC values. Tables 9, 10, and 11 are a comparison of
calculated and reported removal efficiencies. Reported values highlighted in red are values that
are much greater or smaller than the calculated values (+ 5%). It should be noted that the
reported values for TP from the May 31, 2008 and June 3, 2008 event appear to be flipped.
The removal efficiencies are based on EMC values and not mass removal, because there
is a difference between the flow volumes measured in the BaySeparator™ and BayFilter™. The
BayFilter™ flow volume would cause the mass based removal efficiency to be over reported.
36
Table 9 – UMD calculated and reported EMC removal efficiency values for SSC. Reported values were
reported by MASWRC to BaySaver on September 19, 2008. Reported values in red signify an EMC removal
efficiency 5% less than or greater than the EMC removal efficiency calculated by UMD. (,ote: MASWRC
included the April 20, 2008 event, a non-qualifying event, in their average reported change value.)
BaySeparator™ BayFilter™ System
Event Calculated
SSC Change
Reported
SSC Change
Calculated
SSC Change
Reported
SSC Change
Calculated
SSC Change
Reported
SSC Change
11-Apr-08 77% 76% 88% 89% 97% 97%
26-Apr-08 50% 50% 87% 87% 93% 93%
28-Apr-08 57% 58% 72% 71% 88% 88%
31-May-08 38% 38% 89% 89% 93% 93%
3-Jun-08 47% 47% 94% 94% 97% 97%
4-Jun-08 43% 43% 88% 87% 93% 93%
16-Jun-08 -37% -36% 78% 78% 70% 70%
23-Jun-08 -6% -57% 74% 75% 73% 73%
27-Jun-08 42% 39% 87% 87% 92% 92%
9-Jul-08 43% 43% 89% 89% 94% 94%
13-Jul-08 60% 60% 58% 59% 83% 83%
23-Jul-08 45% 45% 62% 62% 79% 79%
2-Aug-08 62% 58% 89% 89% 96% 95%
14-Aug-08 57% 56% 91% 91% 96% 96%
Average
(Prior to 28-Aug-08) 53% 51% 85% 85% 93% 93%
28-Aug-08 51% 84% 92%
25-Oct-08 22% 90% 92%
13-Nov-08 -41%
Not Reported
Yet
66%
Not Reported
Yet
52%
Not Reported
Yet
Overall 47% -- 86% -- 92% --
37
Table 10 – UMD calculated and reported EMC removal efficiency values for TP. Reported values were
reported by MASWRC to BaySaver on September 19, 2008. Reported values in red signify an EMC removal
efficiency 5% less than or greater than the EMC removal efficiency calculated by UMD. (,ote: MASWRC
included the April 20, 2008 event, a non-qualifying event, in their average reported change value.)
BaySeparator™ BayFilter™ System
Event Calculated
TP Change
Reported TP
Change
Calculated
TP Change
Reported
TP Change
Calculated
TP Change
Reported
TP Change
11-Apr-08 26% 26% 22% 21% 43% 41%
26-Apr-08 6% 6% 97% 98% 97% 98%
28-Apr-08 50% 50% 50% 50% 75% 75%
31-May-08 15% 7% 27% 56% 38% 59%
3-Jun-08 8% 15% 56% 26% 59% 37%
4-Jun-08 55% 55% 57% 56% 81% 80%
16-Jun-08 -11% -13% 59% 61% 55% 56%
23-Jun-08 25% 23% 54% 52% 65% 63%
27-Jun-08 3% 2% 83% 83% 83% 83%
9-Jul-08 -1% -2% 70% 70% 70% 69%
13-Jul-08 63% 63% 52% 67% 83% 88%
23-Jul-08 47% 46% 53% 53% 75% 75%
2-Aug-08 55% 53% 77% 79% 90% 90%
14-Aug-08 33% 33% 46% 46% 64% 63%
Average
(Prior to 28-Aug-08) 33% 30% 64% 66% 76% 76%
28-Aug-08 -29% 58% 45%
25-Oct-08 -3% 40% 38%
13-Nov-08 -48%
Not Reported
Yet
37%
Not
Reported Yet
7%
Not
Reported Yet
Overall 20% -- 58% -- 67% --
38
Table 11 – UMD calculated and reported EMC removal efficiency values for turbidity. Reported values were
reported by MASWRC to BaySaver on September 19, 2008. Reported values in red signify an EMC removal
efficiency 5% less than or greater than the EMC removal efficiency calculated by UMD. (,ote: MASWRC
included the April 20, 2008 event, a non-qualifying event, in their average reported change value.)
BaySeparator™ BayFilter™ System
Event
Calculated
Turbidity
Change
Reported
Turbidity
Change
Calculated
Turbidity
Change
Reported
Turbidity
Change
Calculated
Turbidity
Change
Reported
Turbidity
Change
11-Apr-08 0% 2% 70% 70% 70% 70%
26-Apr-08 -12% -12% 79% 79% 76% 76%
28-Apr-08 -44% -37% 46% 46% 22% 26%
31-May-08 -4% -4% 73% 73% 72% 72%
3-Jun-08 9% 9% 66% 66% 69% 69%
4-Jun-08 39% 11% 85% 85% 91% 87%
16-Jun-08 -4% -4% 40% 40% 38% 38%
23-Jun-08 -12% -12% 57% 57% 51% 51%
27-Jun-08 -6% -6% 67% 66% 65% 64%
9-Jul-08 -17% -17% 70% 65% 65% 60%
13-Jul-08 11% 15% 35% 35% 43% 45%
23-Jul-08 3% 3% 32% 32% 34% 35%
2-Aug-08 44% 38% 80% 80% 89% 87%
14-Aug-08 38% 38% 72% 72% 83% 83%
Average
(Prior to 28-Aug-08) 16% 7% 67% 68% 72% 70%
28-Aug-08 -2% 68% 67%
25-Oct-08 10% 61% 65%
13-Nov-08 -30%
Not Reported
Yet
33%
Not Reported
Yet
13%
Not Reported
Yet
Overall 11% -- 64% -- 68% --
6.3.3 Split Samples
6.3.3.1 October 25, 2008
Nine split samples were collected from the October 25, 2008 rainfall event. Three
samples from each of the three sampling locations were split. The samples were analyzed for
SSC, TP, and turbidity; results are presented in Tables 12, 13, and 14. Generally, UMD SSCs
(first half of the sample) were lower, -1 to -34%, than MASWRC SSCs (second half of the
sample). The only exception was sample number 6 from BayFilter™ effluent, which UMD found
to be 95% higher than the value MASWRC found. The UMD and MASWRC TP concentrations
ranged widely, between -66 to 52%. The BayFilter™ number 6 (BFe6) sample was not analyzed
for TP. Turbidity values were within 15% for six of the nine samples, and the other three were
39
between 27 and 68%. Although the percent differences may appear to be large, agreement is
generally satisfactory. The variability in values between UMD and MASWRC is to be expected
with splitting samples in the field, different laboratory equipment, and slight differences in
laboratory handling and analysis.
Table 12 – SSC results of the split samples collected on October 25, 2008.
10/25/2008
Sample ID UMD SSC
(mg/L)
MASWRC
SSC (mg/L)
Difference
(mg/L) Difference
BSi2 15 17 3 -16%
BSi6 125 190 64 -34%
BSi8 335 427 92 -22%
BSe5 163 193 29 -15%
BSe7 108 110 2 -2%
BSe9 432 437 5 -1%
BFe1 29 33 4 -11%
BFe3 17 25 8 -33%
BFe6 25 13 -12 95%
Table 13 - TP concentration results from the split samples collected on October 25, 2008.
10/25/2008
Sample ID UMD TP
(mg/L)
UMD TP
Duplicate
(mg/L)
MASWRC
TP (mg/L)
Difference
(mg/L) Difference
BSi2 0.06 --- 0.04 -0.02 52%
BSi6 0.20 --- 0.26 0.06 -23%
BSi8 0.68 --- 0.91 0.23 -25%
BSe5 0.14 --- 0.21 0.07 -32%
BSe7 0.28 --- 0.23 -0.05 22%
BSe9 0.79 --- 0.80 0.01 -1%
BFe1 0.12 0.15 0.36 0.24 -66%
BFe3 0.12 --- 0.27 0.15 -57%
40
Table 14 - Turbidity results from the split samples collected on October 25, 2008.
10/25/2008
Sample ID
UMD
Turbidity
(,TU)
MASWRC
(,TU)
Difference
(,TU) Difference
BSi2 22 15 -7 45%
BSi6 97 93 -4 5%
BSi8 355 316 -39 12%
BSe5 67 59 -8 13%
BSe7 88 83 -5 6%
BSe9 541 427 -114 27%
BFe1 91 88 -3 3%
BFe3 72 66 -6 9%
BFe6 65 39 -26 68%
6.3.3.2 ,ovember 13, 2008
Nine split samples were collected from the November 13, 2008 rainfall event. Three
samples from each of the three sampling locations were split. The samples were analyzed for
SSC, TP, and turbidity; results are presented in Tables 15, 16, and 17. UMD SSCs (first half of
the sample) were lower, 0 to -85%, than MASWRC SSCs (second half of the sample). The UMD
and MASWRC TP concentrations ranged widely, between -33 to 380%, with no clear
correlation. Turbidity values reported by UMD were generally higher, 4 to 87%, than values
found by MASWRC. Although the percent differences appear to be large, agreement is generally
satisfactory. The variability in values between UMD and MASWRC is to be expected with
splitting samples in the field, different laboratory equipment, and slight differences in laboratory
handling and analysis. The largest abnormality is the TP results from the BSi2 and BSe8
samples, which UMD calculated values three to four times greater than the values reported by
MASWRC. The values UMD found were not typical of the other samples results reported by
MASWRC and UMD, but no bias was determined.
41
Table 15 – SSC results of the split samples collected on ,ovember 13, 2008.
11/13/2008
Sample ID UMD SSC
(mg/L)
MASWRC
SSC (mg/L)
Difference
(mg/L) Difference
BSi2 47 56 9 -16%
BSi4 29 38 9 -25%
BSi8 20 24 4 -17%
BSe2 40 41 1 -3%
BSe4 49 50 1 -2%
BSe8 3 17 14 -85%
BFe2 6 8 2 -22%
BFe6 10 10 0 -4%
BFe10 8 9 1 -12%
Table 16 - TP concentration results from the split samples collected on ,ovember 13, 2008.
11/13/2008
Sample ID UMD TP
(mg/L)
UMD TP
Duplicate
(mg/L)
MASWRC
TP (mg/L)
Difference
(mg/L) Difference
BSi2 0.96 --- 0.20 -0.76 380%
BSi4 0.18 --- 0.14 -0.04 27%
BSi8 0.15 0.17 0.18 0.03 -19%
BSe2 0.19 --- 0.21 0.02 -12%
BSe4 0.24 --- 0.23 -0.01 4%
BSe8 0.92 --- 0.29 -0.63 217%
BFe2 0.13 --- 0.20 0.07 -33%
BFe6 0.15 0.13 0.11 -0.04 33%
BFe10 0.15 --- 0.09 -0.06 67%
42
Table 17 – Turbidity results of the split samples collected on ,ovember 13, 2008.
11/13/2008
Sample ID
UMD
Turbidity
(,TU)
MASWRC
(,TU)
Difference
(,TU) Difference
BSi2 41 27 -13 48%
BSi4 29 19 -9 48%
BSi8 20 11 -9 87%
BSe2 35 26 -9 36%
BSe4 44 30 -14 49%
BSe8 9 7 -2 36%
BFe2 13 12 -1 4%
BFe6 21 12 -9 78%
BFe10 22 12 -9 75%
43
7.0 DISCUSSIO,
7.1 Precipitation
The precipitation data collected at the site demonstrate good agreement with precipitation
data collected in the surrounding area, Figure 14. One discrepancy was the June 27, 2008 event
which original had a value of 0.36 inches reported. That data appeared to be half the actual value,
and the value was adjusted by MASWRC to 0.72 inches. The collection of the precipitation data,
downloaded from the HOBO logger appears to be done in a thorough manner. MASWRC checks
the precipitation data from the site against data from www.wunderground.com (Woodley
Gardens monitoring station) for similarities and differences. The rain gauge appears to be
collecting accurate data, and the data are being checked by MASWRC against additional known
values.
7.2 Flow Balance
The differences in reported flows between the BaySeparator™ and BayFilter™ were the
most troubling aspect of the data being evaluated. Between 4% to 45% of the flow into the
BaySeparator™ was not being recorded in the BayFilter™. The difficulty and uncertainty
associated with area/velocity flow meters has been noted in many other studies. The flows do not
usually balance easily as these test are conducted under field conditions. ETV, in their testing of
the BaySeparator™ in 2003, noted wide ranging differences in reported flows from the influent
to the effluent of the BaySeparator. However, the cumulative volumes being reported by the flow
meters should be in agreement with each other. The two values should balance with 10 to 20%.
Currently, the flow volumes are varying on average by 24%.
The field flow test results show that the BaySeparator™ flow meter may be over
reporting the flow rate and the BayFilter™ flow meter may be under reporting the flow rate,
Tables 4 and 5. Based on the flow test results, the BayFilter™ flow meter was under reporting
flow rates by as much as 50%, and likewise the BaySeparator™ flow meter, at flow rates greater
than 100 gpm, was over reporting by as much as 50%. Also, under high intensity or long
duration rainfall events, the detention system can be completely full and flow is diverted from
the BaySeparator™. This bypass is estimated to have occurred during the August 28, 2008
according to MAWSRC and BaySaver.
44
The final consideration to be taken into account for the flow balance is a break in the
structures or piping. Construction activities have been occurring on site throughout the duration
of the project, and it is possible that damage was done. Notably, the construction activities
occurred in the vicinity of the treatment devices during August and September 2008. Data from
events after those activities had flow balance differences of 9% and 44%.
Although damage to the treatment devices is possible, based on the under and over
reporting of the flow meters in the field flow test, and no clear trend on the percentage of flow
unaccounted for, it seems unlikely that damage is the cause of the discrepancies in flow data.
Most likely the over reporting of BaySeparator™ flow rate and under reporting of BayFilter™
flow rate is the cause of the unbalanced flow volumes.
7.3 Monitoring Results
The EMC data calculated by UMD were found to be similar to the data reported by
MASWRC. Discrepancies in the data are likely due to UMD using mean values of the samples
before and after an uncollected (empty bottle) sample, while MASWRC does not take into
account a sample bottle being empty. A half full or empty sample bottle was found in many of
the events. This was likely due to the flow being below the suction line (approximately 0.5
inches off the bottom of the pipe) or the suction line clogging.
The water quality samples were potentially affected by the lack of sample preservation,
exceeding recommended hold times, and the absence of QA/QC laboratory methods. Sample
preservation has only been used by MASWRC for the October 25 and November 13, 2008
events. Hold times potentially have been exceeded for the samples over the course of the project.
Without laboratory duplicates of TP and turbidity, there are no means to evaluate consistency
between the results being reported. Further QA/QC, both on sample handling and sample
analyses, needs to be a priority, to ensure accurate results.
7.4 Cumulative Mass
Cumulative mass values were reported in the two reports to NJCAT, however these data
are likely to be inaccurate due to the differences in reported flow volumes. If cumulative mass
loadings are to be calculated, the same volume must be used for both, unless flow is bypassing
the detention system.
45
7.5 6on-Qualifying Events
During the course of the field testing some field testing data that were gathered were not
used, because the rainfall or sampling event was determined to be a non-qualifying event. Other
events were not used due to problems with the collection of samples. A few key events are noted
below as to why samples were not collected and/or reported.
The September 6, 2008 event was a high volume rainfall event that caused flooding in the
filter vault. Samples were not collected, because the filter vault auto-sampler capsized and the
BaySeparator™ auto-sampler was jammed.
During the month of May 2008, rainfall events were not sampled, with the exception of
the May 31, 2008 event. During that time period MASWRC was in the process of compiling an
interim field testing data report for MDE.
The month of July 2008 had 3 small rainfall events that occurred from July 1, 2008 to
July 5, 2008 that did not qualify as rainfall events because the precipitation amounts were too
low, less than 0.15 inches.
The April 20, 2008 rainfall event did not meet the requirement of 70% of flow collected.
Flow sampling was not paced properly and there was a time gap between the 24th and 25
th
samples collected.
The September 25, 26, and 30, 2008 rainfall events occurred while construction activities
were occurring on the site.
7.6 Concerns
Over the course of the project an ongoing dialogue has taken place between UMD and
MASWRC regarding the sampling and data evaluation. Discrepancies and deficiencies have
been identified between current practices and the necessary practices. These necessary practices
are based on information in the Standard Methods, the QAPP, and the Tier II protocol. To date
some of these concerns have been addressed, while others remain to be rectified.
46
7.6.1 Concerns Addressed
The concerns that have been addressed are listed below, including what MASWRC did to
solve the problem (listed in italics).
• Sample Labels – The labels needed to be more descriptive. Suggested using color coded
dots, repeat labeling, date, and sample location included on labels. Currently, color coded
dots (one on each side of sample bottle) and a computer generated label with the sample
ID and date are being affixed to the sample bottles.
• Sample Cross Contamination – Potential cross contamination of total phosphorus (TP)
sample due to using same pipette to extract from sample bottle. According to MASWRC,
the pipette is now rinsed between samples sets (i.e., between sampling the BaySeparator
influent and effluent sample sets) and the last sample collected (i.e. #24) is the first
sample being extracted (lowest expected TP).
• Field Log Book – Log book detailing site activities is required (TARP Protocol,
Appendix F, section A9) and use of a log book is also included in QAPP as, “The
samples taken will also be recorded in the field logbook, along with the depth of
precipitation, duration of the storm event, peak rainfall intensity, and peak runoff flow
rate for the event.” Starting with the September 25, 2008 event, MASWRC is using a set
of field checklists as their field log book (see Appendix E).
• Chain of Custody (COC) – A COC is required by TARP (TARP Protocol, section 3.3.5)
and the QAPP states a COC will be used. MASWRC started using a COC for the October
25, 2008 event, Appendix D.
• Sample Preservation is required by TARP and Standard Methods (TARP protocol,
section 3.3.5, SM 4500 and SM 2540) and specified in the QAPP. Starting with the
October 25, 2008 event, samples are placed on ice as soon as they are collected in the
field. Samples are stored at 4 oC (i.e,. in a refrigerator) while awaiting analysis.
• Plastic Bottles – Plastic bottles are being used, which may adsorb phosphate (SM 4500).
Per MASWRC, TP sub-sample is now being taken immediately upon return to the
laboratory.
47
7.6.2 Concerns Remaining
The remaining concerns listed below are related to both the sample handling and analysis
and overall system performance. Potential solutions are listed below also (in italics).
QAPP/TARP Tier II Protocol Requirements
• QAPP Approval – QAPP needed to be approved prior to field testing (TARP Protocol,
section 2.2). A final version of the QAPP was submitted to NJCAT on November 9,
2008. Going forward, the QAPP should be closely followed and any deficiencies between
the current practices and those listed in the QAPP need to be addressed immediately.
• QAPP Requirements – Required sections in the QAPP are identified in TARP Protocol in
Appendix E. The required QAPP sections are derived from EPA Requirements for
Quality Assurance Project Plans (EPA QA/R-5). Examine the TARP Protocol
requirements and the EPA requirements and address all deficiencies between the
requirements and the current QAPP. Possible modifications to the current QAPP may be
necessary.
• Laboratory Accreditation – An accredited/certified laboratory is required for sample
analysis (TARP Protocol, section 3.3.7). MASWRC is currently applying for
accreditation. Continue with the accreditation process, and discuss with 4JCAT the
retroactive application of this requirement.
Laboratory Procedures
• Holding Time – TSS samples should be analyzed within 24 hours because of bacterial
decomposition, and they never should go longer than 7 days without being analyzed (SM
2540). Turbidity samples should be run immediately (SM 2130). Run all samples
immediately after transporting them back to the laboratory. Samples should be collected
from the site within 24 hours of the end of the rainfall event.
• Sample Labels – QAPP states “bottles will be labeled with the sample location (inlet or
outlet), the date and time of the first aliquot (this time will mark the beginning of the
holding time), and the type of sample (flow weighted composite discrete flow
composite).” Could also use samples labels, including sample ID, on sub-samples being
used in the laboratory (i.e., on TSS filter pans and TP digestion beakers).
48
• Sample Preservation – Samples taken for TP analysis should be preserved with sulfuric
acid (SM 4500). Use sulfuric acid to preserve TP samples if analysis is not being done
immediately.
• Plastic Bottles – TP samples should be stored in glass containers if being held more than
24 hours. Transfer TP samples to glass containers if TP analysis is not being done
immediately.
• Consistency – Data analysis revealed a gap in QA/QC and the lack of documentation
regarding field activities (field book) and laboratory analysis. MASWRC personnel should
visit a laboratory to see their protocols and procedures. This would give MASWRC a
better understanding of sample handling and analysis. Create a sample handling and
analysis handbook, which would aid in training new laboratory technicians and have all
personnel confident in their procedures. Keep a laboratory record book that notes all
samples irregularities and why they occurred (i.e., half samples, missing samples, suction
line clogged, filter vault flooded, etc.).
• Sample correlation – Currently, very limited number of laboratory duplicates have been
taken or analyzed, besides SSC, which tends to be lower in the first half of the sample
than the second half of the sample. Conduct more laboratory duplicates to check values
by splitting samples to measure TP and turbidity. Send split samples to a commercial
laboratory to correlate with MASWRC values.
Field Procedures
• Health and Safety Plan– A health and safety plan is required as part of the field testing
(TARP Protocol, section 4), but currently no health and safety plan exists for the field
testing being conducted. The health and safety plan should contain information regarding
confined space entry (the most critical part of the plan), traffic hazards at the site, and
contact numbers in the event of an emergency. The location, including a map, of the
nearest hospital should be included in the health and safety plan. The plan should be with
personnel at the site.
49
Data Reporting
• Volume Balance – The balance in flow between the BaySeparator and BayFilter has not
been closed. The field flow test helped address some of the concerns, and the presence of
an ultrasonic pressure sensor in the filter vault further aids in the understanding of flow
differences. Continue to examine why the flows do not agree within 10 to 20% of one
another. Consider conducting a longer flow test, pulling and checking the flow meters
calibration, and also try to determine if the filter vault or horizontal detention system is
leaking.
• Monitoring Results Reported – The differences in the reported values from the UMD
calculated values most likely are due to flow weighting of half samples. In examining the
data, MASWRC should enact a stricter QA/QC program. The tables are often
cumbersome to use and cannot be checked quickly and efficiently. Keeping all the
information calculated for each sample, in one row, would aid in the checking process.
(i.e., The formulas used to calculate mean concentration can be dragged down, with no
errors introduced through cutting and pasting.)
• Split Samples – The differences between the concentrations UMD found and MASWRC
found were generally minor and could be due to sample handling procedures, sample
preservation, or analyses conducted. The applicable method for each constituent should
be strictly followed, and an intensive QA/QC program implemented, including equipment
blanks, field blanks, field duplicates, baseline (deionized water) samples, and known
value samples.
Field Testing Program
• Loading – The amount of sediment concentration dropped in the summer due to a
decrease in vehicular traffic. Continue the monitoring program for at least one year to
account for seasonal variations.
• Clay Material – There has been an influx of silt and clay material into the system from
the construction activities, which may be impacting the removal efficiencies. Sediment
and erosion controls should be installed on the construction site to prevent erosion onto
the site. (Figures 19 and 20.)
50
Figure 19 - View of construction activities at the site (Courtesy of BaySaver).
Figure 20 - View of sediment washed onto the site. The grassy area was previously
silt and clay with no erosion and sediment controls present.
51
• BayFilter™ Influent – The BaySeparator™ effluent sample is being considered as the
influent to BayFilter. However, a horizontal detention system exists between the
BaySeparator™ effluent and BayFilter™ vault. To properly evaluate the removal
efficiencies of the BayFilter™, samples need to be taken at the influent pipe to the filter
vault, unless the horizontal detention system is considered part of the BayFilter™.
• Oil and Grease Removal – The BaySeparator™ is claimed to have oil and grease
removal, but an auto-sampler (TARP) cannot be used to collect volatile organic samples
(VOCs). A grab sample is needed to analyze oil and grease; once the MASWRC facility is
online, the oil and grease claim should be evaluated.
• Independent testing – BaySaver and MASWRC work in close conjunction on the field
testing of the system. However, the TARP protocol calls for independent third party field
testing of the system. For future studies, BaySaver should provide MASWRC with an
annual budget and allow MASWRC to determine which tests are necessary to provide
accurate data. By ceding control of the budget to MASWRC for the field testing, the
degree of influence that BaySaver could potentially exert over MASWRC would be
limited.
52
8.0 SUMMARY/CO,CLUSIO,
Currently, the MASWRC has qualified samples for approximately 10 inches of
precipitation and 17 qualified events have been sampled. Overall, the field activities being
conducted by MASWRC are in line with acceptable industry practices, except where noted
above in the concerns section. However, the laboratory practices, including sample preservation
and laboratory validation, have been deficient at times. The concerns identified above have been
addressed in many instances, but the remaining discrepancies and deficiencies should be
addressed immediately. The data previously reported should undergo a stringent QA/QC process
to ensure accuracy, and data reported in the future should also be subject to a comprehensive
QA/QC review. Based upon UMD’s evaluation of the MASWRC field testing program, the
EMC and removal efficiency results are within an acceptable range with previous stormwater
field testing programs. There appears to be no significant error or omission on the part of the
sampling and analysis conducted by MASWRC to invalidate the data being collected.
53
REFERE,CES
Al-Hamdan, A.Z., Nnadi, F. N. and Romah, M. S. (2007). “Performance reconnaissance of
stormwater proprietary best management practices.” J. of Env. Science and Health, Part A,
42(4), 427– 437.
American Public Health Association, American Water Works Association, and Water
Environment Federation (1999). “4500-P Phosphorus (SM 4500).” Standard Methods for the
Examination of Water and Wastewater 20th
Edition.
American Public Health Association, American Water Works Association, and Water
Environment Federation (1999). “2540 Solids (SM 2540).” Standard Methods for the
Examination of Water and Wastewater 20th
Edition.
American Public Health Association, American Water Works Association, and Water
Environment Federation (1999). “2130 Turbidity (SM 2130).” Standard Methods for the
Examination of Water and Wastewater 20th
Edition.
BaySaver Technologies, Inc. (BaySaver) (2008). 4JCAT BaySeparator and BayFilter QAPP
Submittal. BaySaver, Mount Airy, MD.
BaySaver. BayFilter System Technical and Design Manual. BaySaver, Mount Airy, MD.
BaySaver. BaySeparator Technical and Design Manual, BaySaver. Mount Airy, MD.
BaySaver (1999). BaySaver®
Separation System Total Suspended Solids Removal Data.
BaySaver, Mount Airy, MD.
BaySaver (2008). Richard Montgomery High School Stormwater Management System #2 Design
Computations, As-Built Conditions, and Test Configuration for MASWRC Testing. BaySaver,
Mount Airy, MD.
54
Carlson, L., Mohseni, O., Stefan, H., and Lueker, M. (2006). Performance Evaluation of the
BaySaver Stormwater Separator System. University of Minnesota – Saint Anthony’s Falls
Laboratory, Minnesota, MN.
Clark, S. and Pitt, R. (2008). “Comparison of Stormwater Solids Analytical Methods for
Performance Evaluation of Manufactured Treatment Devices.” J. of Env. Eng., (134), 259-264.
Clark, S., and Siu, C. (2008). “Measuring Solids Concentration in Stormwater Runoff:
Comparison of Analytical Methods.” Env. Sci. and Tech., (42), 511–516.
John, R., Gray, G., Glysson, D., Turcios, L. M., and Schwarz, G. E. (2000). “Comparability of
Suspended-Sediment Concentration and Total Suspended Solids Data.” USGS Water-Resources
Investigations Report 00-4191, United States Geological Survey.
Liu, B. (2008), Efficiency Assessment of BaySeparator and BayFilter Systems in the Richard
Montgomery High School –Preliminary Test Report. Mid-Atlantic Stormwater Research Center,
Mount Airy, MD.
Maryland Department of the Environment (MDE) “TMDL.” TMDL,
<http://www.mde.state.md.us/Programs/WaterPrograms/TMDL/home/index> (Nov. 10, 2008).
MDE. Stormwater Program Fact Sheet.
<http://www.mde.state.md.us/Programs/WaterPrograms/SedimentandStormwater/index.asp>
(Nov. 8, 2008).
New Jersey Corporation for Advanced Technology (NJCAT) (2004). 4JCAT Technical
Verification. NJCAT, New Jersey.
National Oceanic and Atmospheric Agency. “Climate of Maryland.”
<http://www5.ncdc.noaa.gov/climatenormals/clim60/states/Clim_MD_01.pdf> (Sept. 3, 2008).
The Technical Acceptance Reciprocity Partnership (2003). The TARP Protocol for Stormwater
Best Management Practice Demonstrations.
55
Washington State Department of Ecology, Water Quality Program (2008). Technology
Assessment Protocol Ecology (TAPE) Guidance for Evaluating Emerging Stormwater Treatment
Technologies.
United States Environmental Protection Agency (USEPA). “Clean Water Act.” 4PDES,
<http://cfpub.epa.gov/npdes/cwa.cfm> (Nov. 10, 2008).
USEPA. “NPDES history.” 4PDES, <http://www.epa.gov/region01/npdes/history.html> (Nov.
10, 2008).